RS Edge for Loans & Structured Products: A Data Driven Approach to Pre-Trade and Pricing
The non-agency residential-mortgage-backed-securities (RMBS) market has high expectations for increased volume in 2020. Driven largely by expected changes to the qualified mortgage (QM) patch, private-label securities (PLS) issuers and investors are preparing for a 2020 surge. The tight underwriting standards of the post-crisis era are loosening and will continue to loosen if debt-to-income restrictions are lifted with changes to the QM patch.
PLS programs can differ greatly. It’s increasingly important to understand the risks inherent in each underlying pool. At the same time, investment opportunities with substantial yield are becoming harder to find without developing a deep understanding of the riskier components of the capital structure. A structured approach to pre-trade and portfolio analytics can help mitigate some of these challenges. Using a data-driven approach, portfolio managers can gain confidence in the positions they take and make data influenced pricing decisions.
Industry best practice for pre-trade analysis is to employ a holistic approach to RMBS. To do this, portfolio managers must combine analysis of loan collateral, historical data for similar cohorts of loans (within previous deals), and scenarios for projected performance. The foundation of this approach is:
- Historical data can ground assumptions about projected performance
- A consistent approach from deal to deal will illuminate shifting risks from shifting collateral
- Scenario analysis will inform risk assessment and investment decision
RiskSpan’s modeling and analytics expert, Janet Jozwik, suggests a framework for analyzing a new RMBS deal with analysis of 3 main components: deal collateral, historical performance, and scenario forecasting. Combined, these three components give portfolio managers a present, past, and future view into the deal.
Present: Deal Collateral Analysis
Deal collateral analysis consists of: 1) a deep dive into the characteristics of the collateral underlying the deal itself, and 2) a comparison of the collateral characteristics of the deal being analyzed to similar deals. A comparison to recently issued deals can highlight shifts in underlying collateral risk within a particular shelf or across issuers.
Below, RiskSpan’s RS Edge provides the portfolio manager with a dashboard highlighting key collateral characteristics that may influence deal performance.
Example 1: Deal Profile Stratification
Example 2: Deal Comparative Analysis
Past: Historical Performance Analysis
Historical analysis informs users of a deal’s potential performance under different scenarios by looking at how similar loan cohorts from prior deals have performed. Jozwik recommends analyzing historical trends both from the recent past and from historical stress vintages to give a sense for what the expected performance of the deal will be, and what the worst-case performance would be under stress scenarios.
Recent Trend Analysis: Portfolio managers can understand expected performance by looking at how similar deals have been performing over the prior 2 to 3 years. There are a significant number of recently issued PLS that can be tracked to understand recent prepayment and default trends in the market. While the performance of these recent deals doesn’t definitively determine expectations for a new deal (as things can change, such as rate environment), it provides one data point to help ground data-driven analyses. This approach allows users to capitalize on the knowledge gained from prior market trends.
Historical Vintage Proxy Analysis: Portfolio managers can understand stressed performance of the deal by looking at performance of similar loans from vintages that experienced the stress environment of the housing crisis. Though potentially cumbersome to execute, this approach leverages the rich set of historical performance data available in the mortgage space.
For a new RMBS Deal, portfolio managers can review the distribution of key features, such as FICO, LTV, and documentation type. They can calculate performance metrics, such as cumulative loss and default rates, from a wide set of historical performance data on RMBS, cut by vintage. When pulling these historical numbers, portfolio managers can adjust the population of loans to better align with the distribution of key loan features in the deal they are analyzing. So, they can get a view into how a similar loans pool originated in historical vintages, like 2007, performed. There are certainly underwriting changes that have occurred in the post-crisis era that would likely make this analysis ultra–conservative. These ‘proxy cohorts’ from historical vintages can provide an alternative insight into what could happen in a worst-case scenario.
Future: Forecasting Scenario Analysis
Forecasting analysis should come in two flavors. First, very straightforward scenarios that are explicitly transparent about assumptions for CPR, CDR, and severity. These assumptions-based scenarios can be informed with outputs from the Historical Performance Analysis above.
Second, forecasting analysis can leverage statistical models that consider both loan features and macroeconomic inputs. Scenarios can be built around macroeconomic inputs to the model to better understand how collateral and bond performance will change with changing economic conditions. Macroeconomic inputs, such as mortgage rates and home prices, can be specified to create particular scenario runs.
How RiskSpan Can Help
Pulling the required data and models together is typically a burden. RiskSpan’s RS Edge has solved these issues and now offers one integrated solution for:
- Historical Data: Loan-level performance and collateral data on historical and pre-issue RMBS deals
- Predictive Models: Credit and Prepayment models for non-agency collateral types
- Deal Cashflow Engine: Intex is the leading source for an RMBS deal cashflow library
There is a rich source of data, models, and analytics that can support decision making in the RMBS market. The challenge for a portfolio manager is piecing these often-disparate pieces of information together to a cohesive analysis that can provide a consistent view from deal to deal. Further, there is a massive amount of historical data in the mortgage space, containing a vast wealth of insight to help inform investment decisions. However, these datasets are notoriously unwieldy. Users of RS Edge cut through the complications of large, disparate datasets for clear, informative analysis, without the need for custom-built technology or analysts with advanced coding skills.